Slides available here!


Prof. Andreas Herkersdorf, TU Munich, Germany


Conquering MPSoC Design and Architecture Complexity with Bio-Inspired Self-Organization


The complexity of multi-objective hardware/software system design became the key challenge for the effective deployment of multi-/manycore processor platforms in embedded and high performance computing systems. Whether we struggle with MPSoC power, resilience, mixed-criticality, debugging and/or security shortcomings, these are all consequences of the demand-driven rapid increase in function integration and corresponding overall system complexity. Nature has powerful mechanisms to cope with the inherent complexity of living organisms, e.g. the visceral nervous or immune system, which are dominantly founded on the principle of emergent self-organization: a hierarchy of system constituents follow few, individual and simple rules which manifest in an overall desirable system behavior (or its opposite, chaotic behavior, which usually has being eliminated by nature through evolution). These mechanisms inspired initiatives like autonomic and organic computing.
My talk will show examples for the successful applicability of self-organization on homogeneous multicore processors for optimizing workload balancing, power consumption and system resilience. Evolution, i.e. the identification of efficient local rules and behavior, is being emulated by means of generic online and hardware-assisted reinforcement machine learning techniques. As outlook, I would like to discuss with MPSOC Forum participants the suitability of the presented approach for MPSoC diagnosis / software debugging and, in general, opportunities as well as limitations for broader adoption.


Andreas Herkersdorf is a professor in the Department of Electrical Engineering and Information Technology and also adjunct to the Department of Informatics at Technische Universität München (TUM). He received the Dipl.-Ing. degree from TUM in 1987 and the Dr. degree from ETH Zurich, Switzerland, in 1991, both in electrical engineering. Between 1988 and 2003, he has been in technical and management positions with the IBM Research Laboratory in Rüschlikon, Switzerland. Since 2003, Dr. Herkersdorf is director of the Chair for Integrated Systems at TUM. He is a senior member of the IEEE, member of the DFG (German Research Foundation) Review Board and serves as editor for several journals on design automation, communications electronics and information technology. His research interests include application-specific multi-processor architectures, IP network processing, Network on Chip, system level SoC modeling and design space exploration methods, and self-adaptive fault-tolerant computing.

* If you wish to modify any information or update your photo, please contact the web